> ## Documentation Index
> Fetch the complete documentation index at: https://docs.sisera.xyz/llms.txt
> Use this file to discover all available pages before exploring further.

# Sisera AI Copilot, Takes & Macro Intelligence

> Conversational market analysis, structured stock assessments, forkable Takes, macro regimes, and live social feeds.

**Sisera AI** (`/copilot` and `POST /v1/copilot/chat`) connects live market data, Pyth Core reference prices, PreStocks private marks, the corroborated event engine, your unified portfolio, and the deterministic policy compiler into a single conversational workspace.

***

## From Question to Executable Thesis

Unlike generic chatbots, Sisera AI classifies operator intent across live system state and transitions naturally from open-ended research to risk-checked execution:

<Steps>
  <Step title="1. Diagnose Fair-Value & Event Drivers">
    > *"Why is OpenAI trading 8% above its issuer mark?"*\
    > *"What changed in NVDAx in the last hour?"*\
    > *"Which Clawpump agents are outperforming their paired stock?"*
  </Step>

  <Step title="2. Stress-Test Bull & Bear Cases">
    > *"Show me the strongest argument for and against buying OpenAI at current levels."*\
    > Calls `POST /v1/stocks/:symbol/assessment` to produce a schema-validated bull case, bear case, valuation context, and catalyst checklist grounded strictly in observed filings and news.
  </Step>

  <Step title="3. Simulate Portfolio Impact">
    > *"What happens to my concentration and stress scenarios if I allocate \$5,000?"*\
    > Evaluates `POST /v1/portfolio/impact` against your current holdings, sector weights, duplicate underlying exposure, and pool exit liquidity.
  </Step>

  <Step title="4. Compile into a Deterministic Policy or Take">
    > *"Buy $2,000 if the premium falls below 3%, liquidity remains above $50,000, and no high-severity negative event appears."*\
    > Compiles directly into a deterministic `TradingPolicy` AST ready for operator review and arming.
  </Step>
</Steps>

***

## Executable "Takes" (`/v1/takes`)

Sisera introduces **Takes**—structured, verifiable market theses that pair a research narrative with live, machine-evaluated entry/invalidation conditions:

* **Compile a Take (`POST /v1/takes/compile`):** Turn a market thesis into an auditable Take bound to a specific asset, target price/premium, and condition set.
* **Fork a Take (`POST /v1/takes/fork`):** Clone another operator's Take into your own workspace and customize its allocation cap, slippage budget, or trigger thresholds.
* **Arm a Take (`POST /v1/takes/arm`):** Convert a Take into an active, 30-second evaluated policy guarded by your account's risk mandate.
* **Tokenize a Take (`POST /v1/takes/tokenize`):** Graduate a high-conviction Take into an autonomous agent or launchable market structure.

***

## Macro Regimes & Live Social Feeds

Alongside company-specific filings, Sisera contextualizes every trade within broader macro and social flows:

<CardGroup cols={2}>
  <Card title="Macro & Chain Regimes (/macro)" icon="globe">
    Tracks interest rate regimes, inflation series (FRED), macro event calendars (`GET /v1/intelligence/macro`), and live cross-chain TVL snapshots (`GET /v1/chains` via DeFiLlama) to classify overall market risk appetite (`GET /v1/macro-regime`).
  </Card>

  <Card title="Real-Time Social Feeds (/social)" icon="rss">
    Aggregates live market chatter from X, Telegram, Reddit, and Discord (`GET /v1/social-feed`), mapping posts onto Catalog assets while explicitly isolating social sentiment from confirmed SEC filings.
  </Card>
</CardGroup>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.